Principles in the Design of Mobile Medical Apps:
Guidance for Those who Care
Kenny R. Lienhard and Christine Legner
University of Lausanne, Faculty of Business and Economics (HEC), Department
of Information Systems, Lausanne, Switzerland
{kenny.lienhard,christine.legner}@unil.ch
Abstract. The promises of mobile technology in healthcare have led to a great
many mobile apps in public app stores that target patients with specific
illnesses. Medical experts have criticized the status quo of mobile medical apps
owing to the low level of professional medical involvement in mobile app
design, leading to weak clinical performance and a poor integration of these
tools into clinical practice. Grounded in an action design research study, we
build and evaluate a mobile app for elderly patients with age-related macular
degeneration. We formalize our learnings and provide a set of design principles
to guide the effective and feasible construction of mobile medical apps. Our
study systematically develops design knowledge that helps to bridge the current
gap between the rapid advances in mobile technology and the specific needs of
the healthcare sector.
Keywords: mobile health, mobile medical app, mobile patient monitoring,
healthcare, design science
1
Introduction
Mobile applications (mobile apps) are seen as a potentially transformative technology
that provides individual level support to healthcare consumers and new ways for
physicians to partner and interact with their patients. While information used for
personal healthcare is traditionally captured via self-reporting surveys and doctor
consultations, mobile devices with embedded sensors offer opportunities to establish a
continued exchange of information between patients and physicians. Such
information exchanges are particularly important concerning patients with chronic
illnesses. The promises of mobile health and cutting-edge mobile technologies have
led to a great many mobile health apps in public app stores. The two largest mobile
platforms, Android and iOS, host more than 165,000 mobile apps on medical topics,
of which 9% address topics of screening, diagnosis and monitoring a broad spectrum
of illnesses [1]. We focus on this category, and refer to them as mobile medical apps.
Innovation in mobile technology has outpaced the critical evaluation of the impact
of mobile medical apps [2, 3]. Medical experts have criticized the current state of
mobile medical apps, because they are predominantly technology-driven and thus fail
to meet the requirements of clinical practice [2]. While existing mobile apps over13th International Conference on Wirtschaftsinformatik,
February 12-15, 2017, St. Gallen, Switzerland
Lienhard, K.R.; Legner, C. (2017): Principles in the Design of Mobile Medical Apps: Guidance for Those
who Care, in Leimeister, J.M.; Brenner, W. (Hrsg.): Proceedings der 13. Internationalen Tagung
Wirtschaftsinformatik (WI 2017), St. Gallen, S. 1066-1080
1066
emphasize technological aspects, they tend to poorly integrate medical expertise [4].
For instance, only 32% of 63 mobile apps that target patients with colorectal diseases
were built based on medical professionals’ input on development and content [5]. In
another study, a mobile app for melanoma risk assessment available on public app
stores was compared with an experienced physician’s risk assessment for low-risk and
high-risk lesions [6]. The researchers found that the mobile app’s sensitivity (i.e. its
ability to correctly identify those with a disease) was 20% and its specificity (i.e. its
ability to correctly identify those without the disease) 92%. The low involvement rate
of medical professionals not only increases the risk that ineffective or even potentially
harmful tools will be used by patients, it also leads to a poor acceptance rate of mobile
medical apps among physicians and therefore a low integration of these tools into
daily clinical practice [7]. Based on these studies, we argue that integrating software
developers’ and medical professionals’ expertise is a prerequisite for designing
successful mobile medical apps. However, we still lack knowledge about how to best
design mobile medical apps that integrate both aspects. We address this gap in the
research with the following research question: What are suitable principles in mobile
medical app design?
Our research follows a design science research approach “that uses artifact design
and construction (learning through building) to generate new knowledge and insights
into a class of problems” [8]. In a 30-month action design research (ADR) [9] study,
we worked in an interdisciplinary team of researchers and practitioners to develop a
mobile medical app for elderly patients with low vision due to cases of age-related
macular degeneration. Age-related macular degeneration is a form of sight loss
caused by damage to the retina. The result is a shadowlike void in the center of a
patient’s visual field. We evaluated our artifact in two clinical studies with 124
patients and continuously reflected on learnings, which allowed us to generate design
knowledge throughout the process and to refine insights when new inputs became
available. We have formulated and transformed our learnings into a generic set of
design principles that guide the effective and feasible construction of mobile medical
apps. Here, we systematically develop design knowledge that helps to bridge the
current gap between the rapid advances in mobile technology and the creation of
sustainable mobile app solutions in the healthcare sector.
The paper proceeds as follows. In Section 2, we provide a background on the
literature of our study, identify the gap in the research, and formalize the research
question. We then describe the research method. Next, we present Alleye – a mobile
app that targets patients with age-related macular degeneration. In Section 5, we
formalize the learnings from our project and present a set of principles to guide
mobile medical app design. Finally, we summarize our paper’s contributions, describe
the research limitations, and provide an outlook for future research.
2
Background
Mobile medical apps leverage modern mobile devices such as smartphones and tablet
computers, built-in sensor technology, and related software development kits (SDKs)
1067
to screen, diagnose, and monitor a patient’s illness. Screening is the routine
examination of individuals for indications of illness or of high risk for illness [10].
Diagnosis is the inferred state that an illness is present in a person [10]. Mobile
patient monitoring uses “technology to manage, monitor, and treat a patient’s illness
from a distance” [11] once an illness has been attributed to an individual.
A number of studies have developed mobile app solutions for particular illnesses,
including diabetes [12, 13], asthma [14], and depression [15], and have reported
lessons learnt. Goyal et al. [13] take a user-centered design approach, ensuring that
the features of a mobile app are informed by the needs of patients with type 2
diabetes. The resulting application allows patients to self-monitor their physical
activities, diets, and weights, to identify glycemic control patterns in relation to their
lifestyles, and to guide them towards remedial decision-making. Årsand et al. [12]
illustrate that their mobile app can motivate type 2 diabetes patients to think about
how they can improve their health. The authors conclude that their system has the
potential to support the collaboration between patients and clinicians. In another
study, Oresko et al. [16] integrate a Holter monitor with mobile technology and
develop smartphone-based cardiovascular disease detection. Another study presents a
remote monitoring system for elderly patients with multiple chronic conditions [17]
that allows users to see current medical reports on their smartphones based on sensor
data, to perform new measurements, and to communicate with caregivers via the
mobile app. Schnall et al.’s study evaluates existing mobile apps for patients living
with HIV and concludes that the design of such mobile apps requires a thoughtful,
patient-centered, and evidence-based approach [18].
From a medical perspective, recent healthcare research has revealed that a large
number of mobile apps available in public app stores are not based on empirical
evidence [19]. These shortcomings can have serious consequences. For instance, Wolf
et al. [20] measure the performance of four mobile apps that evaluate photographs of
skin lesions. When such a picture is evaluated, the mobile app gives the user feedback
about the likelihood of malignancy. The sensitivity of the investigated mobile apps
ranged from 6.8% to 98.1%. Hamilton and Brady [4] link the weak performance of
some existing mobile apps to low professional medical involvement in the design of
mobile apps. Based on the analysis of 111 mobile apps that focus on pain
management, one study found that the content of mobile apps contain misleading
claims and a lack of academic references [7].
Software technology-oriented communities suggest mobile health frameworks that
target developers of mobile apps [21, 22]. One study provides an ontology-based
context model and a related application framework that focuses on alarm notification
in chronic patient care [22]. Broens et al. [21] suggest a framework to facilitate the
use of contextual information (i.e. context acquisition, context provisioning, and
context reasoning) for user-tailored mobile apps. While mobile health frameworks are
valuable to ensure software component re-usability or functional decomposition, they
target the project’s implementation phase rather than the design and conceptualization
of a mobile medical app. Other studies suggest specific architectural approaches for
mobile health app usage [23]. For instance, Kumar et al. [24] performed a
comprehensive survey on the use of ubiquitous computing for remote cardiac patient
1068
monitoring. They discuss the architecture and quality of service characteristics of the
underlying platform for mobile cardiac monitoring systems.
We identified two research opportunities in the scientific literature. First, current
studies on mobile medical apps mainly report on the design of illness-specific tools
without abstracting higher-level concepts or design principles from their specific
solutions. This makes it difficult to apply the learnings from one illness-specific study
to the design of a mobile medical app in another context. Second, the rapid advances
in mobile technology outpace the rigorous and critical evaluation of the impacts of
mobile apps. This leads to a situation in which mobile medical apps continue to
proliferate, with little evidence of their effectiveness and little support for
understanding how best to design these tools [2]. We build on these gaps in the
research and seek to answer the following research question: What are suitable
principles in mobile medical app design?
3
Research Method
Considering our research goal, we opted for the design science research paradigm,
which emphasizes a construction-oriented view of information systems (IS), i.e.
research centered around designing and building innovative information technology
(IT) artifacts to solve the identified business needs [25]. The research we present here
derives rigor from the effective use of the medical and the IS knowledge bases. Our
research process followed the ADR approach proposed by Sein et al. [9], a research
method for “generating prescriptive design knowledge through building and
evaluating ensemble IT artifacts in an organizational setting” (see Figure 1). The
ADR project we present here was an engaged research collaboration between
academics (two medical and two IS researchers) and practitioners (two senior
physicians, a graphic designer, and a software developer). Our ADR study seeks to
build and evaluate an innovative IT artifact (i.e. a mobile medical app) and uses
heuristic theorizing to synthesize information about artifact solution [26]. The ADR
project started in November 2013 with the problem formulation and continued with
two building, intervention, and evaluation cycles. The project reached its first
complete state in May 2016, when we formalized our learnings from constructing the
IT artifact into a set of design principles.
Problem formulation: Our research was driven by the practical need to design a
mobile medical app that provides a way for patients to participate in the identification
of age-related macular degeneration and the monitoring of this illness. Our case is
particularly interesting from both a medical and an IS perspective. From a medical
perspective, global projections of any age-related macular degeneration cases are 196
million by 2020, rising to 288 million in 2040 [27]. Current treatment regimens in
age-related macular degeneration are suboptimal, owing to 1) the late identification of
treatable age-related macular degeneration, 2) non-individualized treatment leading to
under-treatment in about 30% of patients, and 3) the challenge to identify the best
time for re-treatment after successful treatment and/or treatment suspension. From an
IS design perspective, the patients are elderly people who are usually not familiar
1069
with touch-based mobile devices; also, owing to their limited vision, it becomes
particularly challenging to design an easy-to-use mobile app. Our research effort can
be classified as IT-dominant [9], since it emphasizes creating an innovative
technological design as its outcome. Our review of extant literature about mobile
medical apps revealed that existing studies often report on the design of illnessspecific tools, which makes it difficult to transfer their findings when constructing an
artifact in another context.
1. Problem formulation
•
•
•
Inspired by practice: develop
a mobile medical app (IT
artifact) for patients with agerelated macular degeneration
Literature reports on illnessspecific mobile medical app
design without abstracting
higher-level design principles
Little evidence of mobile
medical apps’ effectiveness
and efficiency
3. Reflection and learning
4. Formalization of learning
•
•
•
2. Building, intervention, and
evaluation (BIE)
•
•
Month 0 – 20, 1. BIE cycle:
alpha version of IT artifact,
formative evaluation within
clinical study, 17 participants,
SUS score: 77
Month 20 – 30, 2. BIE cycle:
beta version of IT artifact,
summative evaluation within
clinical study, 107 participants,
SUS score: 85
•
Systematic review on mobile
medical apps: usability
evaluation and clinical
effectiveness and efficiency
Compare project findings to
the results of the literature
review
Transfer project-related
learnings to the broader class
of problems (mobile medical
apps)
•
•
Generalize problem instance
to the design of mobile
medical apps
Generalize solution instance
from our case into a class of
solution
Formulate and articulate
learnings, derive design
principles
Figure 1. Action Design Research Process (following Sein et al. [9])
Building, intervention, and evaluation (BIE): We built the IT artifact in an agile
software development approach with short iterations. During the building phases, we
conducted regular meetings (approx. every second week) to evaluate current
prototypes in the project team. Each meeting led to new requirements and guided the
development of the next prototype version, which we then distributed to members of
the ADR team. Overall, we created more than 50 prototypes. Collected data (e.g. field
notes from meetings, emails, and visualizations of graphic user interfaces) informed
the design of the next prototype version. We also used the prototypes for research
purposes, since the data provided us with a history of errors and learnings from the
building phases. During the first BIE cycle, we built and evaluated an alpha version of
the mobile medical app that included all the functionalities with a strong focus on the
measurement task. At that stage, we did not prioritize the presentation and
interpretation of the test results in way that is easy to understand by patients. After a
project duration of 18 months, we engaged in a naturalistic formative evaluation to
determine areas for improvement and refinement of our alpha version. The clinical
study took place at the Department of Ophthalmology of the Cantonal Hospital of
1070
Lucerne – Switzerland’s largest public eye clinic. This study, in which 17 patients
took part, was approved by the Ethics Committee and allowed us to get first-hand
feedback from potential end-users. Each patient provided written informed consent.
All patients who had had an ophthalmological consultation at the hospital’s eye clinic
between June and October 2014 were evaluated for inclusion. Patients were excluded
if they were unable to use the mobile medical app, i.e. owing to cognitive or visual
problems. Patients had a mean age of 78.1 years, and the proportion of women was
53%. Of the participants, 13% used a smartphone daily, while 60% had never used
one. During the clinical study, the measurement task had to be executed three times
per week, between two routine clinical visits (approx. one month). After data
cleaning, we had 240 measurement results.
In the second BIE cycle, we implemented the learnings from the first BIE cycle to
build a beta version of the mobile app. The beta version’s aim was to be selfexplanatory to patients, which required a complete re-design of the user interface and
easy-to-understand communication of the measurement results. After several
iterations and new prototype versions, we performed a second clinical study as a
naturalistic summative evaluation of the IT artifact. We enrolled 107 patients in this
study. All patients who had had an ophthalmological consultation at the hospital’s eye
clinic were evaluated for inclusion. Participants provided written informed consent,
and the study was approved by the Ethics Committee. This time, the patients had a
mean age of 72.8 years, and the proportion of women was 47%. Of these participants,
40% used a smartphone daily and 53% had never used a smartphone. Compared to the
first clinical study, we did not only provide iPod touch devices with the mobile app
pre-installed; patients could also use their own smartphone and could download the
mobile app via the Apple app store. Of the patients, 29 used their own device. The
participants performed repeated self-monitoring measurements between monthly
ophthalmological examinations. We collected more than 4,500 measurement results
over this clinical study.
In both clinical studies, patients provided oral feedback on the mobile app’s userfriendliness, answered pre-defined questions of interest, and filled out the System
Usability Scale (SUS). Originally developed by Brooke [28], the SUS is a valid and
reliable tool for measuring usability [29]. The SUS has received much attention in the
scientific community and, after 30 years from its initial presentation, “has certainly
stood the test of time” [29]. In view of the fact that the study participants were elderly
patients with substantial visual impairment, and sometimes also impaired cognitive
functions, we opted for the short, straightforward SUS as an evaluation instrument.
Based on the 500 tools investigated as reference, a SUS score higher than 68 is
considered to be above-average [30]. Since the participants in our clinical study were
native German speakers, we relied on a German translation of the SUS that was made
available via a crowdsourcing project [31].
Reflection and learning: We conducted reflection and learning cycles in parallel to
the two BIE cycles. To get an overview of the current state of mobile medical apps,
and to learn about their clinical effectiveness and efficiency, we collaborated with
medical researchers to perform a comprehensive literature review that integrated the
medical and the IS perspectives. We continuously consulted the collected data from
1071
the ADR project, i.e. the history of prototypes, presentation material, emails,
questionnaire results, opinions from study participants, and measurement results. This
allowed us to compare our findings to the literature review results and to apply our
situated learning to findings that apply to a broader class of problems. The outcome of
the reflection and learning stage was a preliminary set of design principles.
Formalization of learning: Following Sein et al. [9], we distinguish between three
levels for this conceptual move. First, we generalized the problem instance to the
design of mobile apps that support the monitoring and screening of specific illnesses.
Second, we generalized the solutions instance into a class of solution, abstracting
highly specific solutions concepts from our own ADR project to make the concepts
applicable to the entire class of problems. Third, we captured the knowledge gained in
developing an illness-related mobile medical app. Building on the design principles
we identified and refined in the previous stage, we fully formulated and articulated
our learnings. In the derivation of the design principles, we followed the heuristic
theorizing framework suggested by Gregory and Muntermann [26].
4
A Mobile Medical App for Age-related Macular Degeneration
In an engaged academic-practitioner relationship, we built and evaluated Alleye – a
mobile app that seeks to provide a way for patients to participate in screening and
monitoring age-related macular degeneration. We created this mobile app for Apple’s
iPhone and the iPod touch, targeting iOS 8.0 and later. It builds on HTML5
technologies and is wrapped inside a web view that provides access to native platform
features such as a camera and secure storage. The mobile app has four components:
instructions, setup, measurement, and feedback (see Figure 2). For the instructions,
Alleye includes a help-center with visual graphics explaining all the functionalities.
These graphics were also used by the research assistant to explain the mobile app’s
use during clinical studies, and patients received a printed booklet with large visual
graphics and brief explanations. During initial setup, the patient must insert a unique
identification code, so as to match measurement results with his or her electronic
health records at the eye clinic. Patients choose to perform the measurement task in
training mode or in test mode. The measurement task implemented in Alleye is based
on a computerized version of a Vernier hyperacuity alignment task. Hyperacuity is a
property of our visual system that allows us to see straight lines as straight. The term
derives from the fact that it detects misalignments of borders with a precision that is
up to 10 times better than visual acuity. In Alleye, we implemented an alignment task
that examines the extent to which the visual system is capable to see straight lines as
straight. The performance of an individual completing the measurement task
empirically measures the hyperacuity level. Owing to the fact that a drop in
hyperacuity precedes a drop in visual acuity, Alleye is capable to detect a decrease of
visual acuity before a person sees less. Since this task is monocular (i.e. only the
treated eye is open), the patients must select an eye (left or right) before they begin to
measure. At the end of each measurement, the patient confirms that the measurement
is valid (e.g. that there have been no disturbances). Feedback is provided right after
1072
the patient successfully completes a measurement task. The feedback indicates a score
and a color scheme inspired by traffic lights, comparing the score to previous test
results.
The alpha version of Alleye’s user interface followed generic practical guidelines
for mobile development. During evaluation in a real-life medical setting, 17 patients
reported an average SUS score of 77. From a medical perspective, the first clinical
study revealed that the measurement task (i.e. the assessment of the hyperacuity level)
is a promising instrument for screening and monitoring patients with age-related
macular degeneration. However, we encountered some issues with the mobile app
design. For instance, patients could hardly read written text, had difficulties with
insufficient color contrast, and the navigation (based on a standard icon-based menu)
was unclear to them. To re-design the user interface during the build phase of the
second BIE cycle, we simulated the look of the user interface for patients with limited
vision and a shadowlike void in the center of their visual field by holding a filter over
the mobile device. This provided important insights into the use of colors, minimum
font size, or the number of words that should appear on one screen. Further, we
implemented a very structured navigation (i.e. minimizing variability via limited
navigation options) with buttons occupying the entire screen width in order to guide
patients along the mobile app’s four components (instruction, setup, measurement,
and feedback). The complete redesign of the user interface during the build phase of
the second BIE cycle did not impact the measurement task, and captured clinical data
could still be compared between the two clinical studies.
Figure 2. Screenshots of Alleye: Instructions, Setup, Measurement, and Feedback
The second clinical study sought to test the mobile app’ use over several months in a
real-world medical setting. This would help us to gather data for a longitudinal
clinical study and to learn whether or not patients were willing to adopt Alleye and its
re-designed user interface. Of 107 patients, 83 provided oral feedback on userfriendliness and filled out the SUS. Compared to the first BIE cycle, the SUS score
increased from 77 to 85. We corrected for age and frequent smartphone use in the
SUS analysis. In the unadjusted analysis, the estimated mean increase in the SUS
score was 8.2 (95% CI: 1.3 to 15.1), while in the adjusted analysis the estimated mean
1073
increase in the SUS score only dropped slightly (average SUS score increase 7.5, 95%
CI: 0.5 to 14.4, p = 0.036). Thus, the user interface optimized for patients with agerelated macular degeneration created a higher SUS score. This score was confirmed
by oral feedback from patients and a decrease in the help needed by the research
assistant to explain the mobile app’s usage.
5
Principles in Designing Mobile Medical Apps
Based on the insights gained from the BIE cycles and reviews of other studies on
mobile medical apps, we derived four design principles. The design principles’
purpose and scope are to provide guidance on how to design mobile medical apps.
They capture the knowledge gained about the development of Alleye, and formalize
this knowledge to guide the design of other instances of the same class. Justificatory
knowledge represents insights from the literature that inform, explain, and validate
our design decisions. Table 1 summarizes the design principles, exemplary
instantiation based on our research project, and justificatory knowledge.
Table 1. Design Principles and Exemplary Instantiation
Design principles
Instantiation in Alleye
DP1: Mobile medical apps
should consist of four
functional components that
guide a patient: instruction,
setup, clinical measurement,
and analysis and feedback.
•
•
•
•
DP2: The user interface
should be adapted to cope
with patients’ physical and
cognitive restrictions.
• Limited vision: high color contrast, large
font sizes, and buttons that occupy screen
width
• Cognitive restrictions: limited navigation
options, simplify medical information
[32, 33]
DP3: A mobile medical app
should build on a robust
medical knowledge base,
ensuring an evidence-based
approach to mobile app
design.
• Measurement: hyperacuity level informs
diagnosis of age-related macular
degeneration
• User interface and the use of sensor
technology remain independent of the
medical knowledge base
[4, 7, 20]
DP4: Mobile medical apps
should facilitate both
patients’ and physicians’
routines.
• Instructions trigger conversation between
the patient and the physician
• Patient feedback should be aligned with
actions that the physician can manage and is
integrated into his or her routines
[34, 35]
Instruction: help center in mobile app
Setup: identification, select eye for testing
Measurement: alignment task
Analysis and feedback: score and color
scheme
1074
Justificatory
knowledge
[23]
Design principle 1 (DP1): Mobile medical apps should consist of four functional
components that guide a patient: instruction, setup, clinical measurement, and
analysis and feedback. From our experiences with Alleye, we found that mobile
medical apps have four stable functional components that are specific for patient
screening and monitoring. Given the logical order of these components, primary
navigation elements of mobile medical apps should be structured along these four
building blocks in order to provide a patient with guidance. Estrin and Sim [23]
present similar components in their seminal paper on an open mobile health
architecture, but focus more on technical aspects and leave out the instruction
component. The latter is important because it assists the patient through the mobile
app’s usage and provides explanations on the setup, the clinical measurement task,
and the provided feedback. Instructions might be communicated via graphics, audio,
and/or videos that support patients who have difficulties reading text. The setup
component is crucial to prepare the measurement and to provide context-specific
information relevant to perform and analyze the measurement task. For instance,
personal data such as weight, age, clinical data, or configuration options might modify
the measurement task, are important factors to interpret measurement results, or have
diagnostic value (i.e. modify the probability of the presence of the illness). The
illness-related measurement is key to the mobile medical app. The analysis and
feedback component provides the patient with information about his or her
measurement results and might suggest context-specific actions such as the advice to
contact a physician. At the same time, feedback should motivate the patient to
continue performing clinical measurements. For instance, Marin et al. [36] suggest
serious games as a means to keep patients engaged.
Design principle 2 (DP2): The user interface should be adapted to cope with
patients’ physical and cognitive restrictions. In the development of Alleye, we
experienced the limitations of general human interface guidelines provided by mobile
platform providers to design mobile apps that target their operating systems. These
guidelines provide various user interface patterns, which are helpful to design mobile
apps that target a broad audience. However, as we have seen with Alleye, the same
human interface guidelines do not consider the very specific limitations of patient
groups such as impaired vision, cognitive impairment, or limited motor functions.
Designers of mobile medical app should bear in mind end-users’ physical and
cognitive restrictions [32, 33]. For instance, patients with Parkinson’s disease might
have difficulties entering data via a smartphone’s tiny keyboard. Thus, Parkinson’s
patients could be provided with structured forms and large buttons to enter data. On
the other hand, patients with poor vision can be provided with audio guides and
speech recognition instead of written guidance and text fields to enter data. For
building the user interface, it is helpful to simulate end-user limitations, so that the
designer feels how the mobile app’s form and functions works in the hands of future
users. In our project, the physicians guided the implementation of this simulation to
ensure that designers work with scenarios that are realistic to real-life occurrences. In
the case of Alleye, studies on user interface design for elderly people informed the
mobile app’s development. However, usability studies have certain limitations, since
they focus on a user’s physical limitations rather than on cognitive restrictions. With
1075
Alleye, we have seen that presenting correct, unbiased information that is hardly
understandable by patients not only renders this information useless, but can also
cause misunderstandings in patient-physician communication. An evaluation with
potential users was required to ensure that patients have the cognitive capabilities to
understand the information communicated via the mobile medical app.
Design principle 3 (DP3): A mobile medical app should build on a robust medical
knowledge base, ensuring an evidence-based approach to mobile app design. A
robust medical knowledge base builds trust among physicians [4, 7], laying the
ground for an implementation of mobile medical apps in clinical practice. In Alleye,
the assessment of the hyperacuity level informs the diagnosis of age-related macular
degeneration. Our mobile app builds on this principle, which is robust. In the course
of the use of Alleye, the medical knowledge base might inform us that the changes in
hyperacuity are not the same in the various retinal conditions. It might also inform us
that, besides hyperacuity, other easily accessible parameters can be measured. While
the availability of new sensors embedded in mobile devices might facilitate or even
enable the measurement of additional signs and symptoms, they do not necessarily
impact the biological model. Advances in mobile technology that impact the mobile
app’s underlying medical knowledge base would require the mobile app to be
clinically reassessed, ensuring that the approach to mobile app design remains
evidence-based. While it is very likely that mobile medical apps’ forms and functions
require adaption and optimization over time owing to technological advances, the
underlying biological models and particularly the manifestations (signs and
symptoms) of an underlying illness remain fairly stable over time. This is crucial from
a medical perspective. Only the stable measurement of clinical parameters allows
medical researchers to perform longitudinal clinical studies, and physicians in the
hospital can compare a patient’s test results over a certain timeframe.
Design principle 4 (DP4): Mobile medical apps should facilitate both patients’ and
physicians’ routines. Patients potentially have a long-term relationship with their
physicians. In our project, instructions within Alleye were designed with a specific
purpose: They should allow a physician to explain to the patient the mobile app’s
usage within a few minutes. After basic instructions, the mobile app’s use should be
self-explanatory. With Alleye, we also learnt that feedback provided after the
measurement tasks should be aligned with actions that can be handled by a medical
practice. For instance, if the mobile app asks a patient to contact his or her physician
because his or her measurement task scores are decreasing, the physician in the
medical practice must be aware of the meaning of this call to action. Mobile medical
apps offer unique opportunities to improve the quality of the patient-physician
relationship [34], since they allow for a continued exchange of clinically useful
information that might have remained unrevealed during a routine consultation; such
exchanges are of great importance for chronic illnesses in particular. Thus, the design
of a mobile medical app for patients is not something that can be done in isolation.
Any design of patient routines is linked to the design of the physician’s clinical
routines. To be fully implemented in routine patient care, it is crucial that a mobile
medical app considers requirements from clinical practice. This impacts especially
two functional components: instruction (trigger of a patient’s routine) and feedback
1076
(call to action at the end of a patient’s routine). On the one hand, physicians need to
explain to the patient the mobile app’s use (instructions). This new task should be
implemented in existing clinical routines. On the other hand, data collected via a
mobile app should become part of the physician’s decision-making process. Thus, the
mobile app might ask a patient to contact their physician for consultation if their
measurement results worsen. The integration of mobile medical apps into clinical
routines faces several hurdles, including a lack of knowledge or training on mobile
medical apps or incompatibility with current healthcare practices and technology
platforms [34]. It is only when physicians adapt their clinical routines to a patient’s
use of mobile medical apps that the technology’s potential can be fully explored.
While DP1 identifies the functional components that guide a patient in using a
mobile medical app, DP2 to DP4 provide specific insights on a mobile medical app’s
architectural design. The identified design principles foster a patient-physician
relationship, ensuring that both the patient’s and the physician’s requirements are
addressed when designing a mobile medical app.
6
Conclusion
Mobile medical apps continue to proliferate, with little evidence of their clinical
effectiveness and efficiency. From a bottom-up perspective, a number of studies have
focused on illness-specific cases of mobile medical apps (e.g. [13–15]). While these
studies have revealed important illness-specific insights on building mobile apps in
their specific domain, their findings were hardly generalizable to other illnesses. From
a top-down perspective, more software technology-oriented approaches provided
abstract mobile health frameworks (e.g. [21, 22]) that can be implemented in a broad
range of mobile medical apps. This paper links these two research streams by
generalizing the solution instance (i.e. a mobile app that targets patients with agerelated macular degeneration) into a class of solution. Thus, we abstracted highly
specific solution concepts from our project to make the concepts applicable to the
entire class of problems (i.e. a mobile medical app that targets patients with a specific
illness). Therefore, our study is among the very first to provide principled design
knowledge for mobile medical apps. The suggested design principles form a
theoretical contribution [26], since they extend the body of knowledge on the creation
of mobile app solutions in the healthcare sector. The design principles also assist
practitioners in solving current and anticipated problems in the design of mobile
medical apps. The abstraction of our learnings allows practitioners to build on the
suggested design principles and to apply them in the development of a mobile medical
app that targets a different illness than age-related macular degeneration.
Further, our project revealed the importance of involving people from multiple
disciplines in a mobile medical app project. In line with our argument, Nilsen et al.
[37] criticize current mobile health tools that arise from siloed fields with little
reference to previous research. Doing research in an interdisciplinary team involves
additional effort, since the team must create and share a vocabulary. However, such
an interdisciplinary collaboration enables solutions that could hardly emerge within a
1077
single discipline. The application of each of the four suggested design principles in
future studies calls for an interdisciplinary collaboration, since they all require inputs
from both the medical and the software technology’s side.
While our research is grounded in a successful 30-month research project, it has
limitations. Although we integrated our insights with findings from the literature to
inform, explain, and validate our design decisions, we cannot guarantee that our
findings are exhaustive or fully independent of our specific research project. The
research we presented here has opened up possibilities for new and exciting future
research. Our design principles can serve as a basis to develop a design theory, as
suggested by Gregor and Jones [38]. While a design theory provides prescriptions for
the design of an artifact, future research should also study de facto implementation
and how mobile medical apps change interactions between patients and physicians.
What affordances and constraints do mobile medical apps bring to daily clinical
practice? And how does the physician’s corresponding clinical information system
interact with a patient’s mobile medical app? These are important issues to address
the current gap between mobile technology advances and critical evaluation of the
impacts of mobile medical apps in healthcare.
References
1. IMS Institute for Healthcare Informatics: Patient Adoption of mHealth. (2015).
2. Kumar, S., Nilsen, W.J., Abernethy, A., Atienza, A., Patrick, K., Pavel, M., Riley, W.T.,
Shar, A., Spring, B., Spruijt-Metz, D.: Mobile Health Technology Evaluation: The
mHealth Evidence Workshop. American Journal of Preventive Medicine. 45, 228–236
(2013).
3. Boulos, M.N.K., Brewer, A.C., Karimkhani, C., Buller, D.B., Dellavalle, R.P.: Mobile
Medical and Health Apps: State of the Art, Concerns, Regulatory Control and
Certification. Online Journal of Public Health Informatics. 5, (2014).
4. Hamilton, A.D., Brady, R.R.W.: Medical Professional Involvement in Smartphone “Apps”
in Dermatology. British Journal of Dermatology. 167, 220–221 (2012).
5. O’Neill, S., Brady, R.: Colorectal Smartphone Apps: Opportunities and Risks. Colorectal
Disease. 14, e530–e534 (2012).
6. Robson, Y., Blackford, S., Roberts, D.: Caution in Melanoma Risk Analysis with
Smartphone Application Technology. British Journal of Dermatology. 167, 703–704
(2012).
7. Rosser, B.A., Eccleston, C.: Smartphone Applications for Pain Management. Journal of
Telemedicine and Telecare. 17, 308–312 (2011).
8. Kuechler, W., Vaishnavi, V.: A Framework for Theory Development in Design Science
Research: Multiple Perspectives. Journal of the Association for Information Systems. 13,
395 (2012).
9. Sein, M.K., Henfridsson, O., Purao, S., Rossi, M., Lindgren, R.: Action Design Research.
MIS Quarterly. 35, 37–56 (2011).
10. Miettinen, O.S.: Theoretical Epidemiology: Principles of Occurrence Research in
Medicine. Wiley New York (1985).
11. Kay, M., Santos, J., Takane, M.: mHealth: New Horizons for Health Through Mobile
Technologies. World Health Organization. 66–71 (2011).
1078
12. Årsand, E., Tatara, N., Hartvigsen, G.: Mobile Phone-based Self-management Tools for
Type 2 Diabetes: The Few Touch Application. Journal of Diabetes Science and
Technology. 4, 328–336 (2010).
13. Goyal, S., Morita, P., Lewis, G.F., Yu, C., Seto, E., Cafazzo, J.A.: The Systematic Design
of a Behavioural Mobile Health Application for the Self-management of Type 2 Diabetes.
Canadian Journal of Diabetes. (2015).
14. Burnay, E., Cruz-Correia, R., Jacinto, T., Sousa, A.S., Fonseca, J.: Challenges of a Mobile
Application for Asthma and Allergic Rhinitis Patient Enablement – Interface and
Synchronization. Telemedicine and e-Health. 19, 13–18 (2013).
15. Burns, M.N., Begale, M., Duffecy, J., Gergle, D., Karr, C.J., Giangrande, E., Mohr, D.C.:
Harnessing Context Sensing to Develop a Mobile Intervention for Depression. Journal of
Medical Internet Research. 13, e55 (2011).
16. Oresko, J.J., Jin, Z., Cheng, J., Huang, S., Sun, Y., Duschl, H., Cheng, A.C.: A Wearable
Smartphone-based Platform for Real-time Cardiovascular Disease Detection via
Electrocardiogram Processing. Information Technology in Biomedicine, IEEE
Transactions on. 14, 734–740 (2010).
17. Boulos, M.N., Wheeler, S., Tavares, C., Jones, R.: How Smartphones Are Changing the
Face of Mobile and Participatory Healthcare: An Overview, with Example from
eCAALYX. Biomedical Engineering Online. 10, 24 (2011).
18. Schnall, R., Mosley, J.P., Iribarren, S.J., Bakken, S., Carballo-Diéguez, A., Brown III, W.:
Comparison of a User-centered Design, Self-management App to Existing mHealth Apps
for Persons Living with HIV. JMIR mHealth and uHealth. 3, (2015).
19. Majeed-Ariss, R., Baildam, E., Campbell, M., Chieng, A., Fallon, D., Hall, A., McDonagh,
J.E., Stones, S.R., Thomson, W., Swallow, V.: Apps and Adolescents: A Systematic
Review of Adolescents’ Use of Mobile Phone and Tablet Apps that Support Personal
Management of their Chronic or Long-term Physical Conditions. Journal of Medical
Internet Research. 17, (2015).
20. Wolf, J.A., Moreau, J.F., Akilov, O., Patton, T., English, J.C., Ho, J., Ferris, L.K.:
Diagnostic Inaccuracy of Smartphone Applications for Melanoma Detection. JAMA
Dermatology. 149, 422–426 (2013).
21. Broens, T., Van Halteren, A., Van Sinderen, M., Wac, K.: Towards an Application
Framework for Context-aware mHealth Applications. International Journal of Internet
Protocol Technology. 2, 109–116 (2007).
22. Paganelli, F., Giuli, D.: An Ontology-based Context Model for Home Health Monitoring
and Alerting in Chronic Patient Care Networks. In: Proceedings of the Advanced
Information Networking and Applications Workshops, 2007, AINAW’07. 21st
International Conference on. pp. 838–845. IEEE (2007).
23. Estrin, D., Sim, I.: Open mHealth Architecture: An Engine for Health Care Innovation.
Science. 330, 759–760 (2010).
24. Kumar, S., Kambhatla, K., Hu, F., Lifson, M., Xiao, Y.: Ubiquitous Computing for
Remote Cardiac Patient Monitoring: A Survey. International Journal of Telemedicine and
Applications. 2008, 3 (2008).
25. Hevner, A.R., March, S.T., Park, J., Ram, S.: Design Science in Information Systems
Research. MIS Quarterly. 28, 75–105 (2004).
26. Gregory, R.W., Muntermann, J.: Heuristic Theorizing: Proactively Generating Design
Theories. Information Systems Research. 25, 639–653 (2014).
1079
27. Wong, W.L., Su, X., Li, X., Cheung, C.M.G., Klein, R., Cheng, C.-Y., Wong, T.Y.: Global
Prevalence of Age-related Macular Degeneration and Disease Burden Projection for 2020
and 2040: A Systematic Review and Meta-analysis. The Lancet Global Health. 2, e106–
e116 (2014).
28. Brooke, J.: SUS – A Quick and Dirty Usability Scale. Usability Evaluation in Industry.
189, 4–7 (1996).
29. Lewis, J.R., Sauro, J.: The Factor Structure of the System Usability Scale. In: Human
Centered Design. pp. 94–103. Springer (2009).
30. Sauro, J.: Measuring Usability with the System Usability Scale (SUS),
http://www.measuringu.com/sus.php (Accessed: 17.08.2016).
31. Rummel, B.: Crowdsourcing the Translation of SUS,
https://experience.sap.com/skillup/system-usability-scale-jetzt-auch-auf-deutsch/
(Accessed: 29.10.2016).
32. Holzinger, A., Searle, G., Nischelwitzer, A.: On some Aspects of Improving Mobile
Applications for the Elderly. In: Universal Access in Human Computer Interaction. Coping
with Diversity. pp. 923–932. Springer (2007).
33. Demiris, G., Afrin, L.B., Speedie, S., Courtney, K.L., Sondhi, M., Vimarlund, V., Lovis,
C., Goossen, W., Lynch, C.: Patient-centered Applications: Use of Information
Technology to Promote Disease Management and Wellness. A White Paper by the AMIA
Knowledge in Motion Working Group. Journal of the American Medical Informatics
Association: JAMIA. 15, 8–13 (2008).
34. Silow-Carroll, S., Smith, B.: Clinical Management Apps: Creating Partnerships between
Providers and Patients. Issue Brief (The Commonwealth Fund). 30, 1–10 (2013).
35. Free, C., Phillips, G., Galli, L., Watson, L., Felix, L., Edwards, P., Patel, V., Haines, A.:
The Effectiveness of Mobile-health Technology-based Health Behaviour Change or
Disease Management Interventions for Health Care Consumers: A Systematic Review.
PLOS Medicine. 10, e1001362 (2013).
36. Marin, J.G., Navarro, K.F., Lawrence, E.: Serious Games to Improve the Physical Health
of the Elderly: A Categorization Scheme. In: International Conference on Advances in
Human-oriented and Personalized Mechanisms, Technologies, and Services. Barcelona,
Spain (2011).
37. Nilsen, W., Kumar, S., Shar, A., Varoquiers, C., Wiley, T., Riley, W.T., Pavel, M.,
Atienza, A.A.: Advancing the Science of mHealth. Journal of Health Communication. 17,
5–10 (2012).
38. Gregor, S., Jones, D.: The Anatomy of a Design Theory. Journal of the Association for
Information Systems. 8, 312–335 (2007).
1080